Systems and methods for controlling data exposure using artificial-intelligence-based modeling
Systems and methods for controlling the exposure of data privacy elements are provided. The systems and methods may generate an artificial profile model. The artificial profile model may include a constraint for generating new artificial profiles. A signal may be received indicating that a computing device is requesting access to a network location. One or more data privacy elements associated with the computing device can be detected. An artificial profile can be determined for the computing device. The artificial profile may be usable to identify the computing device. The one or more data privacy elements may be automatically modified according to the constraint included in the artificial profile model. The method may include generating a new artificial profile for the computing device. The new artificial profile may include the modified one or more data privacy elements. The new artificial profile may mask the computing device from being identified.
1. A computer-implemented method, comprising:
identifying a set of data privacy elements, wherein a data privacy element characterizes a feature of a computing device, and wherein a data privacy element is detectable by a network host;
receiving a signal indicating that a computing device is requesting access to a network location;
detecting one or more data privacy elements associated with the computing device request to access the network location;
analyzing the set of data privacy elements, wherein an analysis is done using one or more machine learning techniques;
identifying a constraint associated with the set of data privacy elements, wherein the constraint corresponds to one or more constraint dependencies within the set of data privacy elements that indicate a vulnerability;
generating one or more modified data privacy elements that are consistent with the one or more constraint dependencies;
generating an artificial profile, wherein the artificial profile is generated using the one or more modified data privacy elements based on the one or more machine learning techniques, and wherein the artificial profile is usable to identify the computing device; and
establishing access to the network location with the one or more modified data privacy elements of the artificial profile, wherein the artificial profile prevents the one or more data privacy elements from being exposed to the network host.
2. The computer-implemented method of claim 1 , further comprising:
mapping the one or more data privacy elements to a model based on the one or more machine learning techniques, wherein the constraint includes one or more dependencies that map to a correlation between at least two data privacy elements; and
based on the mapping, intelligently predicting other data privacy elements that are related to the one or more data privacy elements included in a previous profile for the computing device.
3. The computer-implemented method of claim 1 , wherein at least two data privacy elements are modified in accordance with the constraint that includes one or more dependencies that map to a correlation between the at least two data privacy elements.
4. The computer-implemented method of claim 1 , wherein the one or more machine learning techniques are unsupervised.
5. The computer-implemented method of claim 1 , wherein the artificial profile is based on a machine learned model, wherein the machine learned model is based on one or more attribution vectors, and wherein an attribution vector represents a detectable characteristic associated with data privacy elements that have been clustered based on a similarity in values.
6. The computer-implemented method of claim 1 , wherein the constraint within the one or more machine learning techniques is automatically learned by data privacy elements that have been dynamically clustered based on a similarity in values.
7. The computer-implemented method of claim 1 , wherein the constraint within the one or more machine learning techniques is defined based on a user-defined rule.
8. The computer-implemented method of claim 1 , wherein the artificial profile obfuscates the computing device when the computing device accesses the network location, wherein the artificial profile modifies at least two data privacy elements in the artificial profile in accordance with the constraint identified by the one or more machine learning techniques.
9. A system, comprising:
one or more data processors; and
a non-transitory computer-readable storage medium containing instructions which, when executed on the one or more data processors, cause the one or more data processors to perform operations including:
identifying a set of data privacy elements, wherein a data privacy element characterizes a feature of a computing device, and wherein a data privacy element is detectable by a network host;
receiving a signal indicating that a computing device is requesting access to a network location;
detecting one or more data privacy elements associated with the computing device request to access the network location;
analyzing the set of data privacy elements, wherein an analysis is done using one or more machine learning techniques;
identifying a constraint associated with the set of data privacy elements, wherein the constraint corresponds to one or more constraint dependencies within the set of data privacy elements that indicate a vulnerability;
generating one or more modified data privacy elements that are consistent with the one or more constraint dependencies;
generating an artificial profile, wherein the artificial profile is generated using the one or more modified data privacy elements based on the one or more machine learning techniques, and wherein the artificial profile is usable to identify the computing device; and
establishing access to the network location with the one or more modified data privacy elements of the artificial profile, wherein the artificial profile prevents the one or more data privacy elements from being exposed to the network host.
10. The system of claim 9 , wherein the operations further comprise:
mapping the one or more data privacy elements to a model based on the one or more machine learning techniques, wherein the constraint includes one or more dependencies that map to a correlation between at least two data privacy elements; and
based on the mapping, intelligently predicting other data privacy elements that are related to the one or more data privacy elements included in a previous profile for the computing device.
11. The system of claim 9 , wherein the one or more machine learning techniques are unsupervised.
12. The system of claim 9 , wherein the artificial profile is based on a machine learned model based on one or more attribution vectors, wherein an attribution vector represents a detectable characteristic associated with data privacy elements that have been clustered based on a similarity in values.
13. The system of claim 9 , wherein the constraint within the one or more machine learning techniques is automatically learned by data privacy elements that have been dynamically clustered based on a similarity in values.
14. The system of claim 9 , wherein the artificial profile obfuscates the computing device when the computing device accesses the network location by modifying at least two data privacy elements in the artificial profile in accordance with the constraint identified by the one or more machine learning techniques.
15. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations including:
identifying a set of data privacy elements, wherein a data privacy element characterizes a feature of a computing device, and wherein a data privacy element is detectable by a network host;
receiving a signal indicating that a computing device is requesting access to a network location;
detecting one or more data privacy elements associated with the computing device request to access the network location;
analyzing the set of data privacy elements, wherein an analysis is done using one or more machine learning techniques;
identifying a constraint associated with the set of data privacy elements, wherein the constraint corresponds to one or more constraint dependencies within the set of data privacy elements that indicate a vulnerability;
generating one or more modified data privacy elements that are consistent with the one or more constraint dependencies;
generating an artificial profile, wherein the artificial profile is generated using the one or more modified data privacy elements based on the one or more machine learning techniques, and wherein the artificial profile is usable to identify the computing device; and
establishing access to the network location with the one or more modified data privacy elements of the artificial profile, wherein the artificial profile prevents the one or more data privacy elements from being exposed to the network host.
16. The computer-program product of claim 15 , wherein the operations further comprise:
mapping the one or more data privacy elements to a model based on the one or more machine learning techniques, wherein the constraint includes one or more dependencies that map to a correlation between at least two data privacy elements; and
based on the mapping, intelligently predicting other data privacy elements that are related to the one or more data privacy elements included in a previous profile for the computing device.
17. The computer-program product of claim 15 , wherein the one or more machine learning techniques are unsupervised.
18. The computer-program product of claim 15 , wherein the artificial profile is based on a machine learned model based on one or more attribution vectors, wherein an attribution vector represents a detectable characteristic associated with data privacy elements that have been clustered based on a similarity in values.
19. The computer-program product of claim 15 , wherein the constraint within the one or more machine learning techniques is automatically learned by data privacy elements that have been dynamically clustered based on a similarity in values.